I would like to train a smallish network using cpu cores in parallel rather than gpu as they are slower.

aaDL.options.ValidationFrequency = 50;

%aaDL.options.ExecutionEnvironment = 'cpu';

aaDL.options.ExecutionEnvironment = 'parallel'; % as an alternative

Answer 1

[Joss Knight]

setenv CUDA_VISIBLE_DEVICES -1

when you first start MATLAB, assuming you are running everything locally.

However, as a general rule you can't gain any benefit training in parallel on the CPU on a single machine. MATLAB is already running everything multithreaded.